Author reply

Cancer chronotherapy consists of the delivery of anticancer agents based on biologic rhythms. In randomized Phase III clinical trials, the incidence of severe oral mucositis was reduced fivefold and that of peripheral sensory neuropathy was halved, whereas antitumor activity was found to be enhanced significantly with chronotherapy compared with constant-rate infusion.1 This treatment method is based on the circadian time dependencies of anticancer agent tolerability and efficacy. Indeed, biochemical, cellular, and molecular rhythms in healthy tissues underlie the large and predictable changes in chemotherapy tolerability. As a result, the maximum tolerated dose can be increased safely and efficacy enhanced. Chronotherapy also addresses the treatment of cancer-associated rhythm alterations with agents targeted at circadian clock impairments.2

In this context, Vincenzi et al. emphasized the potential role of the rhythm of thymidylate synthase (TS) in normal colonic mucosa as a major determinant of fluoropyrimidine chronopharmacology.3 Indeed, TS plays a key role in DNA nucleotide precursor synthesis and represents a critical therapeutic target for 5-fluorouracil (5-FU)-based therapy. A rhythm in TS has been demonstrated for enzyme activity with a maximum near 16:00 hours in human oral mucosa,4 thereby supporting the least toxicity of 5-FU for this tissue after its administration at night.1, 2 However, no rhythm in TS mRNA was found in the same samples.4 We recently reported an 8-hour difference in the peak times of activity and mRNA expression for dehydropyrimidine dehydrogenase, a key enzyme for fluoropyrimidine catabolism, in mouse liver.5 These and other results support that the rhythms in enzymatic pathways relevant for 5-FU cytotoxicity result at least from posttranscriptional regulation. Nevertheless, recent studies have emphasized the clinical importance of TS gene polymorphisms and its relation with TS activity. It has been shown that 3 copies of the tandem repeat TSER in the TS promoter region generated a TS expression level that was approximately 2.6-fold higher than that expressed in subjects with 2 copies of TSER in the TS promotor region.6

It therefore is of great interest to learn from clinicians in a different field that polymorphism of the gene for plasminogen activator 1 (PAI1) accounted for the fact that some individuals demonstrate up to 10-fold circadian variations in circulating PAI1, whereas others display no obvious circadian rhythm. Even more so, the authors conclude that homozygosity for the 4G allele of this gene is associated with increased PAI1 levels only in the morning.7

We agree with Vincenzi et al. that the rhythms in TS activity, protein, and mRNA should be investigated. Let us add that a possible relation between TS gene polymorphism and circadian rhythm should be scrutinized, both in tumor and in healthy tissues. This knowledge could help determine the optimal chronomodulated delivery of fluoropyrimidines to prevent severe intestinal toxicity and to enhance dose intensity. Finally, specification of sampling time for both healthy and tumor tissues could minimize the risk of discordant results in predictive oncology.8